Catalytic Acceptorless Dehydrogenation of Amines with Ru(PR2NR′2) and Ru(dppp) Complexes

Article abstract of DOI:10.1021/acs.organomet.6b00870

[Ru(Cp)(P^Ph₂N^Bn₂)(MeCN)]PF₆ (1; P^Ph₂N^Bn₂ = 1,5-benzyl-3,7-phenyl-1,5-diaza-3,7-diphosphacyclooctane) and [Ru(Cp)(dppp)(MeCN)]PF₆ (2; dppp = 1,3-bis(diphenylphosphino)propane) are both active toward the acceptorless dehydrogenation of benzylamine (BnNH₂) and N-heterocycles. The two catalysts have similar activities but different selectivities for dehydrogenation products. Independent synthesis of a [Ru(Cp)(P^Ph₂N^Bn₂)(NH₂Bn)]PF₆ adduct (3) reveals the presence of a hydrogen bond between the bound amine and the pendent base of the P^Ph₂N^Bn₂ ligand. Preliminary mechanistic studies reveal that the benzylamine adduct is not an on-cycle catalyst intermediate.

Full text of DOI:10.1021/acs.organomet.6b00870

Article
pubs.acs.org/Organometallics
Catalytic Acceptorless Dehydrogenation of Amines with Ru(PR NR′2)
2
and Ru(dppp) Complexes
James M. Stubbs, Richard J. Hazlehurst, Paul D. Boyle, and Johanna M. Blacquiere*
Department of Chemistry, University of Western Ontario, London, Ontario N6A 5B7, Canada
S
* Supporting Information
ABSTRACT: [Ru(Cp)(PPh2NBn2)(MeCN)]PF6 (1; PPh2NBn = 1,5-
2
benzyl-3,7-phenyl-1,5-diaza-3,7-diphosphacyclooctane) and [Ru(Cp)-
(dppp)(MeCN)]PF6 (2; dppp = 1,3-bis(diphenylphosphino)propane)
are both active toward the acceptorless dehydrogenation of benzylamine
(BnNH2) and N-heterocycles. The two catalysts have similar activities
but different selectivities for dehydrogenation products. Independent
synthesis of a [Ru(Cp)(PPh2NBn2)(NH2Bn)]PF6 adduct (3) reveals the
presence of a hydrogen bond between the bound amine and the pendent
base of the PPh2NBn ligand. Preliminary mechanistic studies reveal that the benzylamine adduct is not an on-cycle catalyst
2
intermediate.
Scheme 1. Dehydrogenation of Benzylamine with 1
INTRODUCTION
■
Acceptorless dehydrogenation (AD) and acceptorless dehydro-
genative coupling (ADC) have recently emerged as atom-
economical routes to versatile functionalities such as aldehydes,
esters, carboxylic acids, amides, imines, and amines.1 Generally,
these reactions involve dehydrogenation of an alcohol moiety,
typically followed by nucleophilic attack by another alcohol or
amine molecule. Relatively few catalysts have been reported for
amine dehydrogenation,2 but the reaction represents a low-
waste synthesis of imines that is an alternative to common
oxidative strategies.3 Additionally, release of chemically stored
H2 from amines to give nitriles is desirable for alternative fuel
applications.4 Some of the more successful systems for
acceptorless dehydrogenation are the pincer catalysts developed
by Milstein.1b,5 These systems operate through a cooperative6
H2 removal mechanism that involves proton transfer to the
ligand and hydride transfer to the metal.7 The success of such
through hydrogenation of imine A (termed hydrogen
borrowing10). Catalysis with 1 (3 mol %) was evaluated at
110 °C in a variety of solvents (Table 1). The insolubility of 1
limited its performance in toluene (entry 1), a common solvent
for other2a−d AD catalysts. The polar solvents DMF and DMA
give improved solubility and consumption of BnNH2 (entries 2
and 3), but AD products are not observed and a control
reaction without 1 likewise results in the consumption of
BnNH2. The dominant reactivity is ascribed to a competitive,
uncatalyzed coupling with the solvent. Other high-boiling polar
solvents afford improved product formation (entries 4−6), with
the sustainable11 solvent anisole giving the best performance. A
conversion of 75% is achieved after 2 days, and nearly complete
consumption of BnNH2 is reached after 4 days. This
performance is similar to that of known catalysts2a−c which
reach maximum conversion with similar catalyst loadings (1−5
mol %) and shorter times (ca. 24 h), but at higher temperatures
(115−150 °C). The products generated with 1 are imine A and
nitrile B in a ca. 3:1 ratio, which is distinct from most reported
catalysts that commonly10,12 form hydrogen borrowing product
C, though catalysts for selective production of A or B are
known.2a−c,i Release of the generated H2 under a flow of N2
does not lead to improved conversion or product selectivity.
catalysts inspired us to test the established8 cooperative PR NR′2
2
(1,5-R′-3,7-R-1,5-diaza-3,7-diphosphacyclooctane) ligand fam-
ily. Similar to the case for dehydrogenation, electrocatalytic H2
formation (and the reverse H2 oxidation) is promoted with a
number of Ni, Fe, and Ru complexes, where the pendent amine
of the PR NR′2 ligand acts as an intramolecular base to shuttle
2
protons to/from the metal. Herein, we evaluate the catalytic
performance toward amine dehydrogenation and preliminary
mechanistic details of the known9 [Ru(Cp)(PPh2NBn2)-
(MeCN)]PF6 (1) complex (Scheme 1).
RESULTS AND DISCUSSION
■
Benzylamine (BnNH2) was chosen as the benchmark substrate
that has the three possible dehydrogenation products A−C
(Scheme 1). Imine A is formed following dehydrogenation of
BnNH2 and coupling with a second substrate molecule (also
called transamination), nitrile B is formed through two
successive dehydrogenations, and dibenzylamine C forms
Received: November 20, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Catalytic Optimization for the Acceptorless
a
Dehydrogenation of Benzylamine
b
c
entry [Ru]
solvent
toluene
conversn (%) A (%) B (%) C (%)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
2
2
7
99
71
32
64
76
95
18
94
91
87
6
2
0
0
0
0
1
1
0
3
8
DMF
DMA
THFA
19
22
44
54
69
1
17
10
3
2,4,6-collidine
anisole
20
18
0
d
e
f
anisole
anisole
anisole
34
65
52
50
18
18
0
10
10
10
11
anisole
g
anisole
a
Conditions unless specified otherwise: 250 mM BnNH2, 3 mol %
[Ru], 110 °C, 48 h, in a sealed vial. Quantification was conducted by
calibrated GC-FID using an internal standard, values are an average of
b
two runs, and errors are < 5%. Abbreviations: DMF =
dimethylformamide; DMA = dimethylacetamide; THFA = tetrahy-
Figure 1. Conversion curves for the ADC of BnNH2 (black) with
MeO-ArNH2 under the optimized conditions with catalysts (a) 1 and
(b) 2. Yields, determined by calibrated GC-FID analysis, of reaction
products A (red), B (green), C (purple) and D (blue) are plotted.
Data points represent the average of the two runs, and the error bars
give the span of the conversion values of each data set.
c
d
e
drofurfuryl alcohol. Amount of BnNH2 consumed. 96 h. Substrate
f
g
is C. 100 μL of elemental mercury was added. 15 mol % of NEt3.
Treatment of 1 with amine C gives poor conversion, suggesting
that secondary amines are challenging substrates (entry 8).
Addition of mercury does not negatively affect catalyst activity
(entry 9), supporting the homogeneity of the dehydrogenation
catalyst.
Table 2. Catalytic Acceptorless Dehydrogenation of
a
Benzylamine with Aniline Derivatives R-ArNH2
The noncooperative complex [Ru(Cp)(dppp)(MeCN)]PF6
(2) is also catalytically active toward the dehydrogenation of
b
entry
R
[Ru] conversn (%) A (%) B (%) C (%) D (%)
1
2
3
4
5
6
OMe
1
2
1
2
1
2
98
100
98
15
7
24
10
38
34
32
24
0
58
0
53
8
BnNH2 (Table 1, entry 10; dppp
= 1,3-bis-
(diphenylphosphino)propane). Despite the absence of an
internal base in the ligand backbone, 2 shows very good
conversion (91%) under the optimized conditions. Again the
major product is imine A, but both nitrile B and secondary
amine C are observed as minor products. Thus, an internal base
is not required, suggesting that in the case of 2 the substrate
acts as a suitable intermolecular base. Indeed, addition of NEt3
as an exogenous base for catalyst 2 had no effect on the
performance (entry 11).
H
42
23
73
48
19
34
0
100
98
8
NO2
0
100
24
0
a
Conditions: 250 mM BnNH2, 250 mM R-ArNH2, 3 mol % [Ru], 110
°C, 48 h, in a sealed vial. Quantification was conducted by calibrated
GC-FID using an internal standard, values are an average of two runs,
and errors are < 5%, Conversion curves are included in the
b
Supporting Information R of aniline substrates R-ArNH2.
To further probe the scope and distinction between the
PPh2NBn (1) and dppp (2) catalysts, AD of benzylamine was
2
substrate, catalyst 1 gives the aniline-coupled ADC product D
as the major species with minor amounts of A and nitrile B
(Figure 1a; Table 2, entry 1). A comparison to the reaction of 1
with BnNH2 alone (Table 1, entry 6) shows a similar
distribution of dehydrogenation products B and C. The role
of the aniline is predominantly as a nucleophile to convert the
homocoupled product A to heterocoupled product D. In
contrast, catalyst 2 gives only ca. 10% of D (Figure 1b; Table 2,
entry 2). While the aniline shows minimal participation as a
nucleophile, it dramatically alters the product distribution in
comparison to ADC with BnNH2 alone (Table 1, entry 10).
The Brønsted basicity of MeO-ArNH2 diverts the selectivity of
2 from ADC product A to hydrogen borrowing product C.
With the less nucleophilic aniline H-ArNH2 an unselective
mixture of products is observed for both catalysts 1 and 2
(Table 2, entries 3 and 4). Notably, the dppp catalyst 2 gives
only minor amounts of hydrogen borrowing product C, but the
aniline coupling product D is generated as a major product
(along with nitrile B). This increase in D despite the lower
nucleophilicity of the aniline relative to MeO-ArNH2 is
attributed to the lower Brønsted basicity of H-ArNH2. The
conducted in the presence of para-substituted anilines, R-
ArNH2, to give coupled products D (Scheme 2). In all cases,
Scheme 2. Acceptorless Dehydrogenative Coupling of
Benzylamine with Anilines Catalyzed by 1 or 2
the major product with 1 or 2 after 24 h is the homocoupled
product A (Figure 1 and the Supporting Information). At this
time in all cases, >75% consumption of BnNH2 is observed and
the amount of heterocoupled product D is <10%. Formation of
D at longer reaction times (vide infra) likely proceeds following
nucleophilic attack of the aniline on A, rather than on the
primary imine (PhHCNH) generated after AD of BnNH2. A
comparison of product yields at 48 h reveals distinct selectivity
for the two catalysts 1 and 2 (Table 2). With the MeO-ArNH2
PPh2NBn catalyst 1 mediates ADC in the presence of BnNH2
2
B
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and H-ArNH2 to give A as the dominant product. This
difference in selectivity relative to the reaction with MeO-
ArNH2 is expected on the basis of the lower nucleophilicity of
H-ArNH2, which decreases the yield of D. While proton
shuttling by the aniline cannot be excluded for catalyst 1, it
should be noted that the participation of an external base does
not necessarily preclude a cooperative mechanism for the
PPh2NBn catalyst. Extensive mechanistic studies of [Ni-
catalysts RuH2CO(PPh3)3, RuH2(PPh3)3, and the Shvo catalyst
each give >90% conversion of Ind to indole at a higher catalyst
loading (5 mol %) and higher temperature (165 °C).2g Similar
performance is also found for RuCl2(PPh3)3 under conditions
(2 mol % and 110 °C) that are closer to those used for 1 and
2.14 These prior studies and the results presented here show
little distinction in catalyst performance in the AD of Ind
between established cooperative (i.e., 1 and the Shvo catalyts)
and noncooperative catalysts. However, the PPh2NBn2 catalyst 1
outperforms dppp catalyst 2 in the dehydrogenation of Me-Ind
to give 2-methylindole (Table 3, entries 3 and 4). This suggests
that 1 is more tolerant of steric bulk at the site of
dehydrogenation than 2. Both catalysts show poor performance
in the AD of the six-membered heterocycle 1,2,3,4-tetrahy-
droquinoline (THQ; entries 5 and 6).
(PR NR′22)2]2+ electrocatalysts reveals that a pKa matched
2
external base dramatically improves catalyst performance by
shuttling protons to the correctly positioned pendent amine.13
ADC with NO2-ArNH2 does not give any of the heterocoupled
product D with either catalyst 1 or 2 (Table 2, entries 5 and 6).
The electron-withdrawing nitro moiety decreases the nucleo-
philicity of the aniline sufficiently to inhibit coupling. The
PPh2NBn catalyst gives A and B in a higher yield but similar
The different overall activities and selectivities of PPh2NBn
2
2
ratio (ca. 2.3:1; Table 2, entry 5) in comparison to those
observed without the aniline present (cf. 3:1; Table 1, entry 6).
Catalyst 2 also has similar conversion, but a ca. 15% higher
yield of the hydrogen borrowing product C is found (Table 2,
entry 6) relative to reaction without the aniline (Table 1, entry
10). Overall, the added aniline substrates alter the dehydrogen-
catalyst 1 and dppp catalyst 2 led us to question the role of the
pendent amine of 1 in the dehydrogenation mechanism.
Stoichiometric reactions of 1 were thus conducted to identify
potential catalytic intermediates (Scheme 4). Treatment of 1
Scheme 4. Reactivity of (a) 1 with Benzylamine or
Pyrrolidine and (b) 2 with Pyrrolidine
ation selectivity with both the PPh2NBn (1) and dppp (2)
2
catalysts. The Brønsted basicity of the aniline is a dominant
indicator of selectivity for 2, while the nucleophilic character is
most important for 1.
Complexes 1 and 2 are also competent catalysts for the
acceptorless dehydrogenation of five- and six-membered
heterocycles to give indole and quinoline products (Scheme
3, Table 3). Both catalysts dehydrogenate ca. 90% indoline
Scheme 3. Acceptorless Dehydrogenation of N-
a
Heterocycles by 1 or 2
a
Indoline (Ind), R = H, n = 0; 2-methylindoline (Me-Ind), R = Me, n
= 0; 1,2,3,4-tetrahydroquinoline (THQ), R = H, n = 1.
Table 3. Performance of 1 and 2 toward Acceptorless
a
Dehydrogenation of N-Heterocycles
with 5 equiv of benzylamine at 65 °C does not give catalytic
turnover, but a new product is formed, as judged by the ca. 10
ppm upfield shift of the 31P{1H} NMR signal. In a larger-scale
reaction, the product is isolated (85% yield) and is identified as
amine adduct 3 (Scheme 2a). Benzylamine coordination is
supported by MALDI mass spectrometry, which gives a signal
with an isotope pattern and m/z value (757.2) that match to
simulated values for [3 − PF6 + H]+. The new methylene and
1
aryl signals in the H NMR spectrum overlap with existing
signals, but their presence is evident by a change in integration.
The signal for the amine Ru−NH2Bn moiety is observed at
4.91 ppm, which is ca. 1 ppm downfield in comparison to other
[Ru]-NH2Bn complexes.15 We hypothesize that the downfield
shift may be due to a hydrogen-bonding interaction between
the N−H moiety of the benzylamine ligand and the pendent
tertiary amine of the PPh2NBn2 ligand. Identification of through-
a
Conditions: 250 mM Sub., 3 mol % [Ru], 110 °C, 48 h, in a sealed
vial. Quantification was conducted by calibrated GC-FID using an
internal standard, values are an average of two runs, and errors are <
5%. Conversion curves are included in the Supporting Information.
space interactions from the N−H signal to the methylene of the
1
(Ind) under the optimized catalytic conditions (entries 1 and
2), with a faster rate observed for 1 (see the Supporting
Information for conversion curves). By comparison, hydride
PPh2NBn benzyl moiety by H−1H ROESY NMR analysis are
2
inconclusive due to the overlap of the latter signal with the
methylene of the benzylamine ligand.
C
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Compound 4, the pyrrolidine analogue of 3, was synthesized
Scheme 5. Possible Pathways for the Dehydrogenation of
Benzylamine with Catalyst 1
a
to evaluate the potential for hydrogen bonding between the
metal-bound amine and the pendent amine of the PPh2NBn
2
ligand (Scheme 4a). At the lower temperature used for the
synthesis of 4 (65 °C) relative to catalysis (110 °C), no
1
evidence of dehydrogenated pyrrolidine was observed. H−1H
ROESY analysis of 4 reveals two notable correlations between
one of the PPh2NBn N-Bn substituents and the pyrrolidine
2
ligand: (1) Hs to Hj and (2) Hl to Hv (Figure 2a). These
a
[Ru] = [Ru(Cp)]PF6
Figure 2. (a) Expanded section of the 1H−1H ROESY NMR spectrum
of 4. (b) Thermal displacement plot of 4 (right) with ellipsoids at 50%
probability. Phenyl groups on P1 and P2 and the PF6 anion are
such a route complex 3 would be an on-cycle catalytic species
and a precursor to deprotonation. Thus, it should have the
same, or higher, activity toward amine dehydrogenation in
comparison to precatalyst 1 that must dissociate MeCN prior
to entering the cycle. The noncooperative route is similar,
except an exogenous base (i.e., a second equivalent of
substrate) deprotonates the bound substrate and shuttles the
proton back to the hydride. Finally, proton and hydride can be
transferred to the catalyst through an outer-sphere route (either
concerted or stepwise) without coordination of the amine
nitrogen to the metal center.
−
removed for clarity.
suggest that, in the solution state, the pendent amine is
positioned close to the bound pyrrolidine. In contrast, no
correlation is found between the PPh2NBn2 N-Bn methylene and
the methyl protons of the acetonitrile ligand in 1. The location
of the NH signal for 4 (6.30 ppm) is shifted significantly
downfield relative to related Ru(II) amine complexes (ca. 3−4
ppm)15a,16 and further supports the presence of a hydrogen
bond in solution.
Catalytic testing of 3 under the optimized conditions
revealed that the amine adduct has significantly lower activity
than 1, with only 28% imine formed over 48 h (Scheme 6; see
Single crystals of 4 were successfully obtained, and the
aforementioned intramolecular hydrogen-bonding interaction is
evident from the solid-state structure (Figure 2b). The N1−N3
distance of 2.953(7) Å is in the expected range for similar
intramolecular N−N hydrogen-bonding distances (2.7−3.0
Scheme 6. Catalytic Performance Comparison of
Precatalysts 1 and Benzylamine Adduct 3 toward AD of
Benzylamine
Å).17 The proximal six-membered metallacycle of the PPh2NBn
2
ligand is in a boat conformation, pointing toward the
pyrrolidine ligand. In comparison, the metallacyclic ring in all
crystallized Ru(Cp/Cp*)(PR NR′2)(L) complexes is in a chair
2
conformation with the pendent base pointed away from ligand
L (X = MeCN, Cl, O2), unless the amine is protonated and
hydrogen bonds to L (i.e., N−H···O2).9,18
the Supporting Information for a conversion curve). This
suggests that the benzylamine adduct 3 is not an on-cycle
intermediate and that dehydrogenation does not proceed
through an inner-sphere cooperative mechanism. Instead, 3 is
an off-cycle species that enters the catalytic cycle by amine
dissociation to follow a cooperative outer-sphere pathway or by
cleavage of the hydrogen bond to follow a noncooperative
mechanism, which would be operative for the dppp catalyst 2.
Attempts to synthesize a pyrrolidine adduct with dppp
complex 2 also afforded a new product tentatively assigned as 5
in a 27% yield after 4 h, as judged by 31P{1H} NMR
spectroscopy (Scheme 4b). The product is unstable to isolation
and is accompanied by significant decomposition, as is
evidenced by the formation of solids and a loss of 31P
integration over time. This is further support that a hydrogen
bond is a stabilizing force in amine adducts 3 and 4.
CONCLUSIONS
The catalytic mechanism for 1 could follow one of three
possible general paths: cooperative inner sphere, noncoop-
erative inner sphere, or cooperative outer sphere (Scheme 5).
Amine coordination, to give the isolated compound 3, is the
first step in either a cooperative or noncooperative inner-sphere
pathway. The cooperative route would involve substrate
deprotonation by the pendent base and β-H elimination from
the bound amido. These steps would give a Ru−H bond that
would be protonated by the pendent group to release H2. In
■
The complex [Ru(Cp)(PPh2NBn2)(NCMe)]PF6 (1) is an active
acceptorless dehydrogenation catalyst toward benzylamine, and
it preferentially forms imine and nitrile products. The related
complex [Ru(Cp)(dppp)(NCMe)]PF6 (2) shows competitive
activity, but selectivity favors the hydrogen borrowing product
(Bn2NH). Both catalysts show similar activities, but different
selectivities, toward AD of benzylamine and coupling with
various anilines. They are both competitive catalysts for the
D
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
methanimine, dibenzylamine, and benzonitrile were calibrated relative
to the internal standard (tetrahydronaphthalene).
dehydrogenation of five-membered N-heterocycles. This
comparison of the cooperative PPh2NBn and noncooperative
2
Synthesis of [Ru(Cp)(PPh2NBn2)(benzylamine)]PF6 (3). [Ru-
(Cp)(PPh2NBn2)(NCMe)]PF6 (1) (101 mg, 0.121 mmol, 1 equiv) was
placed in a 100 mL Schlenk flask with a stir bar in the glovebox. Dry
THF (10 mL) and BnNH2 (13 μL, 0.12 mmol, 1 equiv) were added
by micropipet and micro syringe, respectively. The Schlenk flask was
fitted with a condenser and the contents heated to reflux on the
Schlenk line for 4 h. The solvent was removed under vacuum to afford
dppp ligands reveals that product selectivity is the dominant
difference between the catalysts. While the dppp catalyst must
follow a noncooperative pathway, the mode of action of the
pendent amine in 2 is less obvious. Isolation and character-
ization of Ru−benzylamine and Ru−pyrrolidine adducts (3 and
4, respectively) reveal that these species are stabilized by a
hydrogen bond formed with the PPh2NBn2 ligand. Poor catalytic
performance of the benzylamine adduct 3 indicates that it is not
a precursor to substrate deprotonation and is not an on-cycle
catalyst intermediate. This study excludes an inner-sphere
cooperative mechanism for 1, leaving outer-sphere cooperative
and noncooperative mechanisms as possible routes. Since the
aniline basicity in ADC reactions with 1 has minimal effect on
the dehydrogenation selectivity (only the subsequent coupling),
a noncooperative (base-assisted) route is less likely for the
1
a brown powder that was washed with Et2O. Yield: 98 mg (89%). H
NMR (600 MHz, CDCl3): δ 7.64−7.59 (m, Ph-H, 4H), 7.55−7.48
(m, Ph-H, 6H), 7.36−7.28 (m, Ph-H, 6H), 7.25−7.17 (m, Ph-H, 3H),
7.14−7.09 (m, Ph-H, 2H), 7.08−7.03 (m, Ph-H, 2H), 6.94−6.88 (m,
Ph-H, 2H), 4.91 (broad, BnNH2, 2H), 4.73 (s, Cp-H, 5H), 3.66−3.60
(m, NCH2P, NCH2Ph, RuNH2CH2Ph, 8H), 3.47 (s, NCH2Ph, 2H),
3.09 (m, NCH2P, 2H), 2.47 (m, NCH2P, 2H). 31P{1H} NMR (243
1
−
MHz, CDCl3): δ 29.2 (s, RuP), −144.3 (sept, JP−F = 715 Hz, PF6 ).
13C{1H} NMR (151.5 MHz, CDCl3): δ 139.7 (Ph-C ring), 136.5 (Ph-
C ring), 134.2 (Ph-C ring), 134.1 (Ph-C ring), 131.4 (Ph-C ring),
131.2 (Ph-C ring), 130.0 (Ph-C ring), 129.6 (Ph-C ring), 129.1−128.5
(Ph-C ring), 128.4−127.9 (Ph-C ring), 81.1 (s, Cp), 67.4 (s, NCH2Ph)
and 64.7 (s, NCH2Ph), 60.1 (s, NH2CH2Ph), 58.3 (s, NCH2P) and
55.2 (s, NCH2P). MALDI MS (pyrene matrix): calcd m/z 757.2 [3 −
PF6 + H]+, obsd m/z 757.2. A crystalline sample was obtained
following vapor diffusion of Et2O into a concentrated solution of 3 in
acetone. Anal. Calcd for C42H46F6N3P3Ru: C, 56.00; H, 5.15; N, 4.66.
Found: C, 56.47; H, 5.25; N, 4.62.
PPh2NBn catalyst. Elucidation of the dominant pathway in
2
acceptorless dehydrogenation with 1 will be investigated in due
course.
EXPERIMENTAL SECTION
■
General Considerations. All reactions were manipulated under
N2 using standard Schlenk or glovebox techniques unless otherwise
stated. All glassware was oven-dried prior to use. Benzylamine (>98%),
triphenylphosphine oxide (99%), aniline (>99%) and 2,4,6-collidine
(99%) were obtained from Alfa Aesar. Pyrrolidine (>99%) was
obtained from Fluka. NEt3 (99%) was obtained from Caledon
Laboratory Chemicals. Pyrene (98%), anisole (99%), dimethylaceta-
mide (99%), and tetrahydrofurfuryl alcohol (THFA) (99%) were
obtained from Sigma-Aldrich. p-Anisidine (99%) and p-nitroaniline
(99%) were obtained from Oakwood Chemicals. Chloroform-d
(99.8%) was obtained from Cambridge Isotope Laboratories. [Ru-
(Cp)(PPh2NBn2)(NCMe)]PF6, (1) and [Ru(Cp)(dppp)(NCMe)]PF6
(2) were synthesized following literature procedures.9 Dry and
degassed tetrahydrofuran (THF), toluene, dichloromethane (DCM),
hexanes, dimethylformamide (DMF), dioxane, and acetonitrile
(MeCN) were obtained from an Innovative Technology 400-5 Solvent
Purification System and stored under N2. These dry and degassed
solvents, except for MeCN, were stored over 4 Å molecular sieves
(Fluka; activated at 150 °C for over 12 h). Triethylamine was dried
with 4 Å molecular sieves and degassed by bubbing with N2.
Chlorofrom-d was dried with 4 Å molecular sieves and degassed by
bubbing with N2. Benzylamine was dried with NaOH, distilled under
vacuum, and stored under N2. All other chemicals were used as
obtained.
Synthesis of [Ru(Cp)(PPh2NBn2)(pyrrolidine)]PF6 (4). [Ru(Cp)-
(PPh2NBn2)(NCMe)]PF6 (1) (150 mg, 0.180 mmol, 1 equiv) was
placed in a 100 mL Schlenk flask with a stir bar. Dry THF (10 mL)
and pyrrolidine (60 μL, 0.90 mmol, 5 equiv) were added by
micropipet and micro syringe, respectively. The reaction mixture was
heated to reflux on the Schlenk line for 4 h. The solvent was removed
under vacuum to afford a brown product that was washed with Et2O.
Yield: 142 mg (92%). Purity: 90% by NMR. Single crystals were
formed following vapor diffusion of Et2O into a concentrated solution
of product in acetone. Upon dissolving single crystals of 4 in THF or
CDCl3, ca. 10% decomposition is observed by 1H and 31P NMR
spectroscopy in 10−15 min, after which no further decomposition is
observed. The numbering scheme for 4 is given in Figure 3. 1H NMR
Charge-transfer matrix assisted laser desorption/ionization mass
spectrometry (MALDI) data were collected on an AB Sciex 5800
TOF/TOF mass spectrometer using pyrene as the matrix in a 20:1
molar ratio to the complex. Solutions were prepared in DCM and
spotted on a sample plate under an inert atmosphere and transferred
to the instrument in a sealed Ziplock bag. The instrument is equipped
with a 349 nm OptiBeam On-Axis laser. The laser pulse rate was 400
Hz, and data were collected in reflectron positive mode. Reflectron
mode was externally calibrated at 50 ppm mass tolerance. Each mass
spectrum was collected as a sum of 500 shots. All NMR spectra were
recorded on either Inova 400 or 600 MHz or Mercury 400 MHz
1
Figure 3. Numbering scheme for H and 13C NMR assignment for
complex 4.
(600 MHz, CDCl3): δ 7.62 (m, Ha, 4H), 7.53−7.47 (m, Hb, Hc, 6H),
7.36−7.30 (m, Hm, Hn, Hr, Hq, 6H), 7.21 (m, Hl, 2H), 7.13 (m, Hp,
2H), 6.30 (broad, Hs, 1H), 4.72 (s, Cp-H, 5H), 3.76 (s, Hi, 2H), 3.71
(m, N−CHg-P, 2H), 3.70 (s, Hj, 2H), 3.65 (m, N-CHe-P, 2H), 3.23
(m, N-CHg-P, 2H), 2.88 (m, Ht, 2H), 2.63 (m, N-CHf -P, 2H), 2.58
(m, Hu, 2H), 1.76 (m, Hw, 2H), 1.51 (m, Hv, 2H). 31P{1H} NMR (243
1
instrument. H and 13C spectra acquired in CDCl3 were referenced
internally against residual solvent signals (CHCl3) to TMS at 0 ppm.
31P spectra were referenced externally to 85% phosphoric acid at 0.00
ppm. Infrared spectra were collected on a PerkinElmer UATR TWO
FTIR spectrometer. Elemental analysis was performed by Laboratoire
−
MHz, CDCl3): δ 29.3 (s, P-Ph), −144.3 (sept, 1JP−F = 713 Hz, PF6 ).
13C{1H} NMR (151.5 MHz, CDCl3): δ 136.8 (s, Co), 135.2 (s, Ck),
134.0 (dd, 1JC−P = 19.9 Hz, 3JC−P = 19.9 Hz, Cd), 131.3 (m, Ca), 129.9
(s, Cc, Cl, Cp), 126.6 (m, Cb), 129.1 (s, Cq), 129.0 (s, Cm), 128.5 (s, Cr),
128.1 (s, Cn), 81.6 (s, Cp), 66.4 (s, Cj), 65.4 (s, Ci), 62.4 (s, Ct), 58.5
(dd 1JC−P = 26.3 Hz, 3JC−P = 26.3 Hz, Ce), 55.8 (dd′, 1JC−P = 17.7 Hz,
́
́ ́ ́
d’Analyze Elementaire de l’Universite de Montreal. Quantification of
catalytic reactivity was achieved using an Agilent 7890a gas
chromatograph with a flame ionization detector (GC-FID). A HP-5
column was used. Benzylamine, phenyl-N-(phenylmethyl)-
E
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3JC−P = 17.7 Hz, Cg), 26.1 (s, Cw). MALDI MS (anthracene matrix):
micropipet. The solution was transferred to a NMR tube, and an initial
(time = 0) 31P{1H} NMR spectrum was obtained. The tube contents
were transferred back to the vial containing the stir bar, and substrate
(benzylamine or pyrrolidine) (0.5 mmol, 5 equiv) was added. The vial
was stirred and heated to 65 °C in an aluminum heating block for 4 h.
The contents were transferred back into a clean NMR tube and a
31P{1H} NMR spectrum was obtained. If more time points were
obtained, the process of heating in the vial and transfer to NMR tube
was repeated for each subsequent time point.
Attempted Synthesis of [Ru(Cp)(dppp)(pyrrolidine)]PF6 (5).
Complex 2 (77 mg, 0.1 mmol, 1 equiv) was placed in a 100 mL
Schlenk flask with a stir bar, and THF (8 mL) was added. In the
Schlenk flask was placed pyrrolidine (36 mg, 0.5 mmol, 5 equiv). The
contents of the Schlenk flask were stirred and heated to 65 °C for 45 h.
The reaction was monitored over time until all of complex 2 produced
black particles. The solvent was removed under vacuum, and the
31P{1H} NMR spectra were obtained in either proteo-THF or CDCl3,
revealing full decomposition in both solvents.
calcd m/z 717.2 [4 − PF6 − 3H]+, obsd. m/z 717.2. Anal. Calcd for
C39H46F6N3P3Ru: C, 54.17; H, 5.36; N, 4.86. Found for a crystalline
sample: C, 54.61; H, 5.43; N, 4.77.
General Procedure for Catalytic Dehydrogenation Reactions
of Benzylamine. In a glovebox, the following stock solutions were
prepared: benzylamine (322 mg, 3.00 mmol, 1 M) and tetrahy-
dronaphthalene (159 mg, 1.20 mmol, 400 mM) in anisole (3.00 mL);
1 (7.5 mg, 0.011 mmol, 15 mM) in anisole (0.750 mL); 2 (14 mg,
0.019 mmol, 15 mM) in anisole (1.250 mL). Four sets, A−D, of two
vials (eight vials total) containing stir bars were charged with the
benzylamine stock solution (125 μL). In each of these vials was placed
the catalyst stock 1 (250 μL to set A) or 2 (250 μL to sets B and C)
along with additional anisole solvent (125 μL for A−C, 375 μL for D).
Triethylamine (1.1 μL, 0.76 mmol) was placed in each vial in set C.
The final concentrations for vials in sets A−D were 0.25 M in
benzylamine with 3 mol % catalyst loading (A−C), and set D
contained no catalyst. A final vial was charged with substrate/internal
standard stock solution (100 μL) for use as the initial time = 0 (T0)
sample for GC-FID analysis. The vials (except the T0 sample) were
capped and removed from the glovebox and heated to 110 °C with
stirring. After 24 and 48 h one vial from each of the sets was removed
from the heat, cooled, and exposed to air to quench. An aliquot (40
μL) was diluted to 10 mM benzylamine with MeCN (960 μL) and
analyzed by GC-FID. A 20 μL aliquot of the T0 sample was diluted
with solvent (980 μL) and analyzed by GC-FID.
General Procedure for Catalytic Dehydrogenation Reactions
of Benzylamine with Anilines. In a glovebox, the following stock
solutions were prepared: benzylamine (322 mg, 3.00 mmol, 1 M) and
tetrahydronaphthalene (159 mg, 1.20 mmol, 400 mM) in anisole (3.00
mL); aniline (279 mg, 3 mmol, 1 M) in anisole (3.00 mL); 1 (15 mg,
0.22 mmol, 15 mM) in anisole (1.50 mL); 2 (17 mg, 0.022 mmol, 15
mM) in anisole (1.500 mL). Benzylamine and aniline stock solutions
were combined (500 mM). Two sets, A and B, of three vials (six vials
total) containing stir bars were charged with the benzylamine/aniline
stock solution (250 μL). In each of these vials was placed the catalyst
stock 1 (250 μL to set A) or 2 (250 μL to set B). The final
concentrations for vials in sets A and B were 0.25 M in benzylamine
with 3 mol % catalyst loading (A and B). A final vial was charged with
substrate/internal standard stock solution (100 μL) for use as the
initial time = 0 (T0) sample for GC-FID analysis. The vials (except the
T0 sample) were capped and removed from the glovebox and heated
to 110 °C with stirring. After 12, 24, and 48 h one vial from each of the
sets was removed from the heat, cooled, and exposed to air to quench.
An aliquot (40 μL) was diluted to 10 mM benzylamine with MeCN
(960 μL) and analyzed by GC-FID. A 20 μL aliquot of the T0 sample
was diluted with solvent (980 μL) and analyzed by GC-FID.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00870.
Spectral data, diffraction data, and catalysis conversion
graphs (PDF)
Single-crystal diffraction data for 4 (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail for J.M.B.: johanna.blacquiere@uwo.ca.
■
ORCID
Johanna M. Blacquiere: 0000-0001-6371-6119
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by a Natural Sciences and
Engineering Research Council (NSERC) of Canada Discovery
Grant and The University of Western Ontario. J.M.S. thanks
the Ontario Graduate Scholarship program for funding.
General Procedure for Catalytic Dehydrogenation Reactions
of N-Heterocycles. In a glovebox, the following stock solutions were
prepared: indoline (357 mg, 3.00 mmol, 500 mM) and tetrahy-
dronaphthalene (80 mg, 0.60 mmol, 200 mM) in anisole (6.00 mL); 1
(15 mg, 0.022 mmol, 15 mM) in anisole (1.500 mL); 2 (17 mg, 0.022
mmol, 15 mM) in anisole (1.500 mL). Two sets, A and B, of 5 vials
(10 vials total) containing stir bars were charged with the indoline
stock solution (250 μL). In each of these vials was placed the catalyst
stock 1 (250 μL to set A) or 2 (250 μL to set B). The final
concentrations for vials in sets A and B were 0.25 M in indoline with 3
mol % catalyst loading (A and B). A final vial was charged with
substrate/internal standard stock solution (100 μL) for use as the
initial time = 0 (T0) sample for GC-FID analysis. The vials (except the
T0 sample) were capped and removed from the glovebox and heated
to 110 °C with stirring. After 1, 4, 12, 24, and 48 h one vial from each
of the sets was removed from the heat, cooled, and exposed to air to
quench. An aliquot (200 μL) was diluted to 50 mM indoline with
MeCN (800 μL) and analyzed by GC-FID. A 100 μL aliquot of the T0
sample was diluted with solvent (900 μL) and analyzed by GC-FID.
General Procedure for Stoichiometric Probe Reactions with
[Ru(Cp)(dppp)(NCMe)]PF6 (2). Complex 2 (8 mg, 0.01 mmol, 1
equiv) and triphenylphosphine oxide (3 mg, 0.01 mmol, 1 equiv) were
placed in a vial with a stir bar. THF (0.800 mL) was added by
REFERENCES
■
(1) (a) Gunanathan, C.; Milstein, D. Chem. Rev. 2014, 114, 12024−
12087. (b) Gunanathan, C.; Milstein, D. Acc. Chem. Res. 2011, 44,
588−602.
(2) (a) Tseng, K.-N. T.; Rizzi, A. M.; Szymczak, N. K. J. Am. Chem.
Soc. 2013, 135, 16352−16355. (b) He, L.-P.; Chen, T.; Gong, D.; Lai,
Z.; Huang, K.-W. Organometallics 2012, 31, 5208−5211. (c) Prades,
A.; Peris, E.; Albrecht, M. Organometallics 2011, 30, 1162−1167.
(d) Ho, H.-A.; Manna, K.; Sadow, A. D. Angew. Chem., Int. Ed. 2012,
51, 8607−8610. (e) Fujita, K.-i.; Tanaka, Y.; Kobayashi, M.;
Yamaguchi, R. J. Am. Chem. Soc. 2014, 136, 4829−4832. (f) Wu, J.;
Talwar, D.; Johnston, S.; Yan, M.; Xiao, J. Angew. Chem., Int. Ed. 2013,
52, 6983−6987. (g) Muthaiah, S.; Hong, S. H. Adv. Synth. Catal. 2012,
354, 3045−3053. (h) Chakraborty, S.; Brennessel, W. W.; Jones, W. D.
J. Am. Chem. Soc. 2014, 136, 8564−8567. (i) Myers, T. W.; Berben, L.
A. J. Am. Chem. Soc. 2013, 135, 9988−9990.
(3) Largeron, M. Eur. J. Org. Chem. 2013, 2013, 5225−5235.
(4) Grellier, M.; Sabo-Etienne, S. Dalton Trans. 2014, 43, 6283−
6286.
(5) Zell, T.; Milstein, D. Acc. Chem. Res. 2015, 48, 1979−1994.
(6) (a) Khusnutdinova, J. R.; Milstein, D. Angew. Chem., Int. Ed.
2015, 54, 12236−12273. (b) Crabtree, R. H. New J. Chem. 2011, 35,
F
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
18−23. (c) Grutzmacher, H. Angew. Chem., Int. Ed. 2008, 47, 1814−
̈
1818.
(7) (a) Sun, Y.; Koehler, C.; Tan, R.; Annibale, V. T.; Song, D. Chem.
Commun. 2011, 47, 8349−8351. (b) Zhang, J.; Leitus, G.; Ben-David,
Y.; Milstein, D. Angew. Chem., Int. Ed. 2006, 45, 1113−1115. (c) Ben-
Ari, E.; Leitus, G.; Shimon, L. J. W.; Milstein, D. J. Am. Chem. Soc.
2006, 128, 15390−15391. (d) Hou, C.; Zhang, Z.; Zhao, C.; Ke, Z.
Inorg. Chem. 2016, 55, 6539−6551. (e) Cho, D.; Ko, K. C.; Lee, J. Y.
Organometallics 2013, 32, 4571−4576. (f) Yang, X.; Hall, M. B. J. Am.
Chem. Soc. 2010, 132, 120−130.
(8) (a) Bullock, R. M.; Helm, M. L. Acc. Chem. Res. 2015, 48, 2017−
2026. (b) Bullock, R. M.; Appel, A. M.; Helm, M. L. Chem. Commun.
2014, 50, 3125−3143. (c) Liu, T.; DuBois, M. R.; DuBois, D. L.;
Bullock, R. M. Energy Environ. Sci. 2014, 7, 3630−3639.
(9) Stubbs, J. M.; Bow, J. P. J.; Hazlehurst, R. J.; Blacquiere, J. M.
Dalton Trans. 2016, 45, 17100−17103.
(10) Dobereiner, G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681−
703.
(11) Alder, C. M.; Hayler, J. D.; Henderson, R. K.; Redman, A. M.;
Shukla, L.; Shuster, L. E.; Sneddon, H. F. Green Chem. 2016, 18,
3879−3890.
(12) Bui-The-Khai; Concilio, C.; Porzi, G. J. Organomet. Chem. 1981,
208, 249−251.
(13) (a) Raugei, S.; Helm, M. L.; Hammes-Schiffer, S.; Appel, A. M.;
O’Hagan, M.; Wiedner, E. S.; Bullock, R. M. Inorg. Chem. 2016, 55,
445−460. (b) O’Hagan, M.; Ho, M.-H.; Yang, J. Y.; Appel, A. M.;
DuBois, M. R.; Raugei, S.; Shaw, W. J.; DuBois, D. L.; Bullock, R. M. J.
Am. Chem. Soc. 2012, 134, 19409−19424.
(14) Tsuji, Y.; Kotachi, S.; Huh, K. T.; Watanabe, Y. J. Org. Chem.
1990, 55, 580−584.
(15) (a) Nyawade, E. A.; Friedrich, H. B.; Omondi, B.; Mpungose, P.
Organometallics 2015, 34, 4922−4931. (b) Sortais, J.-B.; Pannetier, N.;
Holuigue, A.; Barloy, L.; Sirlin, C.; Pfeffer, M.; Kyritsakas, N.
Organometallics 2007, 26, 1856−1867.
(16) (a) Lummiss, J. A. M.; Ireland, B. J.; Sommers, J. M.; Fogg, D. E.
ChemCatChem 2014, 6, 459−463. (b) Fogg, D. E.; James, B. R. Inorg.
Chem. 1995, 34, 2557−2561.
(17) (a) Dabb, S. L.; Messerle, B. A.; Wagler, J. Organometallics 2008,
27, 4657−4665. (b) Prokopchuk, D. E.; Lough, A. J.; Morris, R. H.
Dalton Trans. 2011, 40, 10603−10608.
(18) (a) Tronic, T. A.; Kaminsky, W.; Coggins, M. K.; Mayer, J. M.
Inorg. Chem. 2012, 51, 10916−10928. (b) Tronic, T. A.; Rakowski
DuBois, M.; Kaminsky, W.; Coggins, M. K.; Liu, T.; Mayer, J. M.
Angew. Chem., Int. Ed. 2011, 50, 10936−10939.
G
DOI: 10.1021/acs.organomet.6b00870
Organometallics XXXX, XXX, XXX−XXX